發(fā)布時(shí)間：2018-03-07 10:07

  本文選題：數(shù)據(jù)驅(qū)動(dòng)建模　切入點(diǎn)：高斯過程模型　出處：《浙江大學(xué)》2016年博士論文　論文類型：學(xué)位論文

【摘要】：隨著當(dāng)今工業(yè)信息化、數(shù)字化進(jìn)程的不斷深入,數(shù)據(jù)驅(qū)動(dòng)建模(data-driven modeling)方法及其應(yīng)用引起了廣泛關(guān)注。高斯過程(Gaussian process, GP)模型通過對(duì)訓(xùn)練數(shù)據(jù)相關(guān)性的分析,可以顯式地給出預(yù)測(cè)值的后驗(yàn)概率分布,從而反映了預(yù)測(cè)值的不確定性。這一特性使得它在處理模型不精確時(shí)的預(yù)測(cè)、控制、優(yōu)化等問題時(shí)具有很大優(yōu)勢(shì)。此外,GP模型本身還具有易于實(shí)現(xiàn)、超參數(shù)自適應(yīng)獲取等優(yōu)勢(shì),逐漸成為了機(jī)器學(xué)習(xí)領(lǐng)域的研究熱點(diǎn)之一,并在過程系統(tǒng)工程領(lǐng)域得到了廣泛應(yīng)用。本文圍繞GP模型在不同類型工業(yè)過程中的建模及相關(guān)應(yīng)用開展研究并取得以下成果：1.利用GP模型所提供的預(yù)測(cè)方差信息,提出了自主動(dòng)的GP模型,建立了更新數(shù)據(jù)的自主篩選策略,使用預(yù)測(cè)方差和預(yù)測(cè)誤差相結(jié)合的方式對(duì)模型預(yù)測(cè)值不準(zhǔn)確的情況作出區(qū)分,以更好的選擇建模數(shù)據(jù)。2.基于GP模型所提供的預(yù)測(cè)分布,提出了一種被稱為“基于自主改進(jìn)GP模型的預(yù)測(cè)控制(AI-GPMPC)"方法,適用于初始模型不準(zhǔn)確情況下的設(shè)定值快速追蹤控制問題。該方法能夠在“搜索當(dāng)前模型所提供的信息”和“探索可能改善控制效果的未知區(qū)域”之間進(jìn)行權(quán)衡,在對(duì)系統(tǒng)輸出進(jìn)行有效控制的同時(shí),通過更新訓(xùn)練集改進(jìn)預(yù)測(cè)模型。3.提出了KL-GP的分布參數(shù)系統(tǒng)建模方法,借助KL分解對(duì)過程進(jìn)行時(shí)空分解和維度縮減,并在各空間維度中分別建立GP模型后,通過時(shí)空合成對(duì)原過程進(jìn)行重構(gòu)獲得輸出預(yù)測(cè)。考慮模型更新的需要,對(duì)KL-GP方法進(jìn)行擴(kuò)展,提出了“自主動(dòng)KL-GP (SA-KL-GP)"的建模方法,利用所得的輸出預(yù)測(cè)方差對(duì)任意時(shí)空點(diǎn)上的建模效果進(jìn)行評(píng)價(jià),并自主選取數(shù)據(jù)以改進(jìn)當(dāng)前模型。為滿足實(shí)時(shí)模型改進(jìn)的要求,提出了改進(jìn)的“迭代選擇KL-GP (RS-KL-GP)"建模方法,利用迭代更新方法減少了更新計(jì)算量。4.針對(duì)間歇過程訓(xùn)練數(shù)據(jù)稀缺導(dǎo)致模型不準(zhǔn)確的問題,提出了一種基于GP模型和期望改進(jìn)進(jìn)行批次間最優(yōu)軌跡的設(shè)計(jì)方法。以不準(zhǔn)確預(yù)測(cè)模型為前提,利用“期望改進(jìn)量”的作為優(yōu)化目標(biāo),該方法可以通過盡量少的試驗(yàn)性生產(chǎn)批次得到最優(yōu)的過程產(chǎn)品質(zhì)量。5.基于過程機(jī)理知識(shí),建立了低壓化學(xué)氣相沉積(LPCVD)過程的仿真研究對(duì)象；使用GP模型建立預(yù)測(cè)模型,利用有限的訓(xùn)練數(shù)據(jù)對(duì)含有空間分布信息的批次過程進(jìn)行建模,用以預(yù)測(cè)空間分布的晶圓表面薄膜厚度。在進(jìn)行優(yōu)化控制的過程中,基于GP模型所提供的預(yù)測(cè)值不確定性,優(yōu)化選擇下一批次的操作變量,以保證過程的穩(wěn)定性。此外,預(yù)測(cè)不確定性也被用于更新GP模型的高效數(shù)據(jù)選擇,盡量減少數(shù)據(jù)采樣,同時(shí)增強(qiáng)模型質(zhì)量。
[Abstract]:With the development of industry information and digitization, data-driven modeling method and its application have attracted much attention. Gao Si process Gaussian process (GP) model is analyzed by analyzing the correlation of training data. The posteriori probability distribution of the prediction value can be given explicitly, which reflects the uncertainty of the prediction value. In addition, the GP model itself has the advantages of easy implementation and super parameter adaptive acquisition, which has gradually become one of the research hotspots in the field of machine learning. And has been widely used in the field of process systems engineering. This paper focuses on the modeling and related applications of GP model in different types of industrial processes, and obtains the following results: 1. Using the prediction variance information provided by GP model, A self-active GP model is proposed, and an independent screening strategy for updating data is established. The inaccurate prediction value is distinguished by combining the prediction variance with the prediction error. Based on the prediction distribution provided by GP model, a method called "predictive control based on autonomous improved GP model (AI-GPMPC)" is proposed. This method is suitable for fast tracking control problems where the initial model is inaccurate. This method can be balanced between "searching for information provided by the current model" and "exploring unknown areas that may improve the control effect". While the system output is effectively controlled, the prediction model is improved by updating the training set. (3) the distributed parameter system modeling method of KL-GP is proposed. The process is decomposed and dimensionally reduced by KL decomposition. After the GP model is established in each dimension, the output prediction is obtained through the reconstruction of the original process by space-time synthesis. Considering the need of model updating, the KL-GP method is extended, and the modeling method of "self-active KL-GP SA-KL-GP" is proposed. In order to meet the requirements of real-time model improvement, an improved "iterative selection KL-GP RS-KL-GP" modeling method is proposed to evaluate the modeling effect at any time and space points by using the output predictive variance obtained, and to select the data independently to improve the current model, in order to meet the requirements of real-time model improvement, an improved" iterative selection KL-GP RS-KL-GP" modeling method is proposed. The iterative updating method is used to reduce the updating computation. 4. Aiming at the problem of model inaccuracy caused by the scarcity of training data in batch process, A design method of optimal trajectory between batches based on GP model and expectation improvement is proposed. Based on the inaccurate prediction model, the "expected improvement quantity" is used as the optimization objective. Based on the knowledge of process mechanism, the simulation object of low pressure chemical vapor deposition (LPCVD) process is established, and the prediction model is established by using GP model. The batch process containing spatial distribution information is modeled with limited training data to predict the thickness of the spatially distributed wafer surface film. In the process of optimization control, the prediction value is uncertain based on GP model. The operation variables of the next batch are optimized to ensure the stability of the process. In addition, the prediction uncertainty is also used to update the efficient data selection of GP model to minimize data sampling and enhance the quality of the model.
【學(xué)位授予單位】：浙江大學(xué)
【學(xué)位級(jí)別】：博士
【學(xué)位授予年份】：2016
【分類號(hào)】：O211.6
，

本文編號(hào)：1578988